Or: In which I am brave, and also in need of feedback.
So, I sat down today and finished the right front half of The Sweater, having finished the back and left front half yesterday. I duly pulled it on over my head and took a look. It wasn't bad, but the side shaping was bothering me, as I thought it would, in that it seemed more dramatic than I'd wanted (it almost looks like there's a pleat in there), and the two stitches between the paired decreases didn't match in width with the two between the increases (because the decreases themselves look as if they belong to that center column, making it appear four stitches wide, while that isn't the case with lifted increases, so those center columns look only two stitches wide). I set it aside and went to eat some lunch and think.
I also grabbed some knitting books to look at while I ate lunch, hoping for inspiration. One of those was Knitting Without Tears, that lovely EZ standby. I'd already been wondering if the right solution mightn't be to snip a stitch and rip out the lower half of the body, something I've never tried before, but of which EZ often speaks so coolly that it seems like a reasonable option. No sooner had I settled down and looked at the cover of the book when it occurred to me (that EZ, inspiring by osmosis) that I'd cleverly left the underarm stitches on holders to pick up when I knit the sleeves; of course, I then had to wonder why I couldn't just drop those stitches right down to the bottom of the increase/decrease section, and reknit it, putting more rows in between the decreases to make the change less drastic, and leaving four stitches between the increases to make the columns look more even. Even though it would mean working back up through a lot of rows, it would still take less time than reknitting the whole body.
Why not, indeed? I only hoped that the ten stitches I'd left on holders would be enough.
When I sat down with the sweater, I realized that, in some prescient moment that I had forgotten, I'd actually put 20 stitches on holders for the underarms, which should more than cover the number of stitches involved in the whole increase/decrease thing. I started by letting ten stitches drop, but once I'd picked them up at the bottom, I realized that I needed more width to cover all of the shaping.
I duly let all twenty stitches drop down to the bottom of the shaping section on the left side and captured them on a dpn.
And I started to work my way back up.
Let me just take a moment here to say that reknitting all of those rows leaves one with quite a bit of time to think, and to think odd thoughts at that. It occurred to me as I was working my way up (and down, and up again), that there's really a lot to be learned from taking one's knitting apart this way and putting it back together. I absolutely understand why those Renaissance artists were so desperate to dissect the human body in order to better paint/draw/sculpt it. There's nothing like looking at the inner workings of something to really understand it. (Please forgive me if this analogy seems gross; I was raised by a nurse and a doctor, and to this day I think Gray's Anatomy is one of the world's coolest books -- I have, in fact, been thinking that it's high time to find a copy for the girls.) I'm starting to think that doing this kind of thing to one's knitting should be part of Knitting 101; even just redistributing the extra unevenness of some of the yarn to neighboring stitches was a really interesting exercise.
The biggest difficulty in the whole project lay in the fact that I was decreasing at a slower rate than I'd done before, which meant that at the point at which my floats were shortest, I was knitting more stitches than I had been on that row the first time I knit it, so things got a bit tight. I worked my way through the decreases, realized that I'd decreased too slowly, and decided to rip back and try again. On that second pass through, I first knit about an inch more than I'd done before, then decreased every three rows, but didn't decrease as many stitches as I'd done when first knitting the body. I then worked my way back up to twenty stitches again, and knitted up the rest of the dropped yarn, remembering to leave four stitches between the increase points.
It's not as smooth and even as my knitting usually is, but I think that the shaping is less drastic, will fit better, and I'm hoping (pleasepleaseplease) that blocking will take care of any residual unevenness.
That's the original on the left (the other side that I still need to do), and the new work on the right. See? Less drastic. I had a picture of the whole sweater, so you could see the difference between the two sides in terms of overall shape, but the camera has taken to turning the occasional picture blue (??), so I can't show you that one.
The feedback part comes in here: do we think (pleaseplease, did I mention?) that blocking will even out those stitches, or do I need to rip the whole body out and start again? I'd rather not, but I also want to be able to wear this. Also, does that shaping (the one on the right) look as much better as I think it does?
The other non-complete-ripping option that comes to mind would be to drop to the bottom of those stitches again, and then to knit them up evenly, staying at 20 stitches the whole way, but switching to smaller and smaller needles as the floats get shorter, and then back up to the right size as they get bigger again. That might actually end up being a pretty neat effect, but I'm not sure if there are hidden pitfalls there that I haven't thought of. Has anyone tried anything like that? I'm starting to become more and more enamored of that idea, actually, as it seems to be such an elegant solution, so someone save me quick if it's not a good one, because I'm now thinking that I'll try it on the second side to see what I think, and then go back and fix whichever side is less good.